This manuscript prepared by Han et al. describes an investigation of secondary organic aerosol (SOA) arising from cyclic volatile methyl siloxanes (cVMs). The SOA yield under a variety of atmospheric conditions was investigated using an oxidative flowtube reactor (OFR). The authors provided detailed discussions on the effect of NOx concentration, presence of seed, and oxidation lifetime. Chemical composition of the resulting SOA was monitored with an aerosol mass spectrometer (AMS). Further, the authors applied the obtained yield in a model to estimate the contribution of cVMS SOA on the global scale. The scope of the study matches that of ACP. The discussion and conclusions of the paper were fully justified, and the literary quality of the manuscript is outstanding. I highly recommend publication in ACP. My comments below can be considered only technical. 

One thing that I feel is missing from the current manuscript is a brief discussion on the mechanism of OH reaction with cVMS. This shouldn't take more than a few sentences or a short paragraph. In addition to H-abstraction on the -CH3 groups, does OH undergo any reaction unique to e.g., O-Si bonds? Pieces of mechanistic information are provided in line 265 - 268, but it is hard for the readers to gauge whether that is generic OH chemistry or something unique to cVMS. 

Figure S6b - It is a little counterintuitive that the Si/O ratio continued to rise during photooxidation. The number of Si in cVMS is fixed, while the molecule should be continuously oxidized. So Si/O should be decreasing as oxidation proceeds? The authors' discussion from line 259 to 270 did not really explain why Si/O was rising continuously. 

Line 151 - the acronym, PTR-ToF-MS is already introduced previously. 

Major Comment

The study was conducted to measure the formation of secondary organic aerosols (SOAs) during OH oxidation of cyclic volatile methylsiloxanes (cVMS) in an oxidation flow reactor (OFR). While the measured data of SOA formation was consistent with other published results obtained similar OFRs, the authors assumed that the results of OFRs would be used to determine the SOA formation in the real atmosphere. However, the assumption was proved incorrect by an earlier study by Charan et al. (2021). Their work was cited in Introduction of the current manuscript, but the main result of the paper was completely ignored (not mentioned/discussed). Charan et al. (2021) supported that the yield of SOA depended strongly on OH concentration and organic aerosol mass concentration (related to initial D5 concentration) among several experimental parameters (e.g., OH concentration, OH exposure, aerosol seed, NOx, initial D5 concentration, reactor type). They also claimed that the formation of secondary aerosols is fairly small (<5% or more likely <1%) under environmentally relevant conditions (i.e., low OH and VMS concentrations). Thus, a large chamber might be a good option (rather than OFRs) to maintain environmental-relevant conditions and a low surface-to-volume ratio. Note that all the SOA yields from OFRs were high and in contrast with those from chamber reactors. However, the current manuscript (by Han et al.) did not discussed any of these important findings. They asserted SOA yield was dependent on OH exposure and applied the OFR results for the real atmosphere. Therefore, without providing a strong result against the conclusions of Charan et al. (2021), the current manuscript was built on an incorrect assumption. Accordingly, the estimated SOA production rate is not reliable.

The calculation of cVMS-SOA concentration (in Eq. 4) is not clear. The equation needs the amount of lost of cVMS, but the quantity was not measured or published. See more in Minor Comment below.

The manuscript states that the number and mass size distribution of aerosols was monitored using a scanning mobility particle sizer (SMPS), but no data was provided and discussed except the statement that the data was used for calculation of effective aerosol densities. The results would be helpful to understand the characteristics of SOA formation under different conditions.

Minor Comment

L51-53: Is this (scarce knowledge) still valid? There have been many studies published since the reference you cited.

L135 & Table S2: Based on my extra calculations, the initial concentration range (i.e., 20-40 ppb & 20-25 ppb in Table S3) is equivalent to ~10,000 times greater than environmental concentrations or similar to the values of Janecheck et al. (2019). See Table S2. The authors should note that the concentrations are high enough to increase potential uncertainty in the extrapolation to the environmental concentrations. This manuscript does not provide individual initial concentrations for each of the cVMS species, but it would be helpful with the data.

L190 (& L29): This study claims that the production of SOA is dependent on OH exposure. However, this is partially true in the limit of OFR experiments. In contrast, Charan et al. (2021) discovered that OH radical concentrations are more important than OH exposure considering all experiment results with OFR and chamber reactors. If the atter is valid, it's better to use OH radical concentrations for the x-axis rather than photochemical age as the initial cVMS concentrations were in a relatively narrow range.

L255-256: This is under unseeded conditions. Please explain why there were more productions under seeded conditions at earlier PAs?

L321-322: The estimation would not be valid based on this study that used OFRs. See Charan et al. (2021) for the detailed discussion.

L323,330: There should be differences between C_cVMS and ΔC_cVMS. I assume SOA concentrations were calculated based on C_cVMS, not ΔC_cVMS. Eq.4 is valid only when C_cVMS = C_in-cVMS and background C_cVMS-SOA=0. Thus, there should be more clarification. In addition, extrapolation should be highly uncertain due to 4 orders-of-magnitude differences of concentrations and overestimated SOA yields.

L332-333: Not clear what values were used in this study. The values were not directly measured.

L617-619: Check this reference: not used.

The manuscript by Han et al describes several cVMS aerosol yield experiments in an oxidation flow reactor (OFR). They use their derived yields to make estimates of cVMS-derived SOA at various locations. cVMS chemistry and aerosol yield is an open question in the field and the results are of interest to the community. However, I have several major comments that I would like to see addressed before I could recommend publication.

Major Comments

• The results of this study need to be discussed in the context of Charan et al., (2021). Charan et al. (2021) found that the aerosol yield for D5 oxidation varied with the OH exposure in an unusual way. I note that when considering the Charan et al., (2021) work, it would be prudent to wait for the ACP version of the manuscript as several points of discussion regarding possible explanations of the differing yields were modified following public discussion. A particular result from Charan et al. (2021) that should be discussed in the context of the current study is that OFR experiments led to higher SOA yields than environmental chamber experiments.
• Further details about the radical chemistry need to be included in the manuscript. While information is given on the ratio or RO2 + NO to RO2 + HO2 reactions in the high NOx experiments, there is no discussion on how prevalent RO2 + RO2 reactions were or even the (not atmospherically relevant) RO2 + OH. Differences in the fate of the peroxy radical in the OFR experiments compared to chamber experiments is one possibility that could explain the Charan et al (2021) results, however, the investigation of RO2 fate was beyond the scope of that paper. I encourage the authors to describe the radical chemistry of these experiments in more detail as it could aid in understanding the apparent discrepancy between OFR experiments and chambers for this chemical system.
• Given the prevalence of cVMS, it is easy for contamination to occur. I would like to see more information on control experiments that were performed to check for contamination and possible impacts on cVMS yields. Additionally, I think the manuscript would benefit from including more examples (SI is ok) of the measurements. In particular, I would like to see some examples of the SMPS data and AMS time series absolute values (not in terms of fraction). The SMPS data in particular should be included as it is central in the yield calculation.
• The back of the envelope calculations for the cVMS SOA loadings need to be better justified. For instance, why is 8.5 days selected for the urban sites? That seems long considering typical transport times. There should also be discussion of the results of Milani et al., (2021) and Pennington et al., (2021) and the consistency or not of these back of the envelope estimates with those studies.

Minor comments

• Figure 2 (and analogous SI figure for high NOx). Given that the cVMS precursors only differ in the number of (C2H6OSi) units, I would have expected to see somewhat more similar mass spectra from their oxidation products. I find it intriguing how different the MS are between the experiments and find some of the ions unexpected. For instance, C12H11O2Si5+ from D5 oxidation is intriguing given the addition of C and loss of O from the parent species. I recognize the electron ionization spectra of complex mixtures can be challenging and non-intuitive to interpret, but I wonder if the authors have considered possible explanations for some of these ions and what they might imply for the identify of oxidation products. Additionally, I wonder if the PTR provides any information on gas-phase oxidation products that is useful.
• Lines 268-270: I would think that continued breaking of Si-O bonds would lead to more volatile products since after the first bond breaking, it would lead to fragmentation and thus smaller molecules.

Technical comments

• Line 70: The conclusion of Alton and Browne, (2020) is more subtle than suggested here. They suggest that while Cl is minor globally, loss locally (for instance in Los Angeles) could be more important given the spatial and temporal overlap of cVMS emissions and sources of Cl (i.e. from ClNO2). This is an important distinction if cVMS yields are high as suggested by this work.
• Lines 84-85: “…indicating necessary conditions when extrapolating SOA yields to the ambient atmosphere.” This sentence does not make sense. Please reword.
• Table S2: Please keep D5 in the same units for all the studies.
• Line 229: Alton and Browne (2020) did not state anything about the relative volatility of different cVMS oxidation products. The only discussion on volatility concerned wall loss and that one of the first-generation oxidation products had a lower vapor pressure than the parent cVMS. Moreover, that vapor pressure is still high enough that it is unlikely to contribute to SOA except at high loadings. Additionally, they did not report D6 results.
• The use of n/z and n/z rather than Si/C and Si/O adds complexity to the paper that I think is unnecessary. I think it would be easier for the reader to understand the results if Si/C and Si/O are used.
• Lines 265-266: This discussion of oxidation as occurring by substitution of one functional group for another is confusing and misleading. I suggest editing to be more clear about the chemistry occurring (OH abstracting a H, etc.).

References

Alton, M. W. and Browne, E. C.: Atmospheric Chemistry of Volatile Methyl Siloxanes: Kinetics and Products of Oxidation by OH Radicals and Cl Atoms, Environ. Sci. Technol., 54, 5992–5999, https://doi.org/10.1021/acs.est.0c01368, 2020.

Charan, S. M., Huang, Y., Buenconsejo, R. S., Li, Q., Cocker III, D. R., and Seinfeld, J. H.: Secondary Organic Aerosol Formation from the Oxidation of Decamethylcyclopentasiloxane at Atmospherically Relevant OH Concentrations, Atmospheric Chem. Phys. Discuss., 1–17, https://doi.org/10.5194/acp-2021-353, 2021.

Milani, A., Al-Naiema, I. M., and Stone, E. A.: Detection of a secondary organic aerosol tracer derived from personal care products, Atmos. Environ., 246, 118078, https://doi.org/10.1016/j.atmosenv.2020.118078, 2021.

Pennington, E. A., Seltzer, K. M., Murphy, B. N., Qin, M., Seinfeld, J. H., and Pye, H. O. T.: Modeling secondary organic aerosol formation from volatile chemical products, Atmospheric Chem. Phys., 21, 18247–18261, https://doi.org/10.5194/acp-21-18247-2021, 2021.